AI Paper Atlas

Research map / 3D vision / Face and shape

Learning Forgery-Aware Lip Representations Without Forgery Priors

Bofan Chen, Hongyu Zhu, Yi He, Sichu Liang, Shi-Lin Wang

CVPR 2026 · 3D vision › Face and shape · approach: Speech-driven 3D facial animation

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
replace rotation pseudo-fakes with mixedfake lips (SML/CML) (ablation)proposed detector · TFG-GRID PersonaTalkAUC (%): 78.27 -> 99.92Improved
replace cross-entropy classifier with asymmetric contrastive loss (ablation)proposed detector · TFG-GRID ER-NeRFAUC (%): 62.83 -> 99.95Improved
replace SupCon with asymmetric contrastive loss (ablation)proposed detector · TFG-GRID PersonaTalkAUC (%): 65.90 -> 99.92Improved
replace SimCLR with asymmetric contrastive loss (ablation)proposed detector · TFG-GRID ER-NeRFAUC (%): 85.83 -> 99.95Improved
add Gaussian distribution matching loss (ablation)proposed detector · AVLips (LOF, IF, OC-SVM, EE)HTER (%): HTER reduced by 0.16 to 3.38 pointsImproved
add human-impostor samples to training (V + HM) (ablation)proposed detector · AVLips human-impostor attackAUC (%): 95.24 -> 99.35Improved
add audio modality (AV variant) (ablation)proposed detector · AVLips avgAUC (%): 96.97 -> 99.84Improved
plug adapter into Siamese and SA-DTH encodersSiamese, SA-DTH · TFG-GRID personalized TFGsAUC (%): up to 40% AUC increaseImproved
train only on real data with no forgery priors vs supervised detectorsproposed detector · TFG-GRID avgAUC (%): 95.26 -> 99.83Improved
default visual-only detector under real-world conditions, human-impostor attacksproposed detector · AVLips human-impostorAUC (%): TD-VSA 99.94 vs 95.24Mixed

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy